I am running a simply one-factor CFA model that includes 4 continuous indicators. I have set the model as follows:
MODEL: !Measurement model MomRel BY mc_1A@1 ms_1A maf_1A ss1_1A;
!Variances mc_1A ms_1A maf_1A ss1_1A; MomRel;
!Means and Intercepts [fc_1A fs_1A faf_1A SS2_1A]; [MomRel];
When running this analysis, I get the following message:
THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE COMPUTED. THE MODEL MAY NOT BE IDENTIFIED. CHECK YOUR MODEL. PROBLEM INVOLVING PARAMETER 12.
Parameter 12 is the ALPHA for my latent variable MomRel, which I believe is its estimated mean. This problem disappears if I add the constraint of [MomRel@0]. I have run multiple 1-factor with 3 to 4 indicators CFAs and I have had the same problem every time, and setting up a constraints for the mean @0 has solved this problem every time.
I know that this type of constraint is used when during CFAs across groups, and I understand that the constraint sets the latent mean to equal zero. Why is this done when doing CFA invariance across groups? And what does it mean that this constraint solves the problems mentioned above in the one-group, one-factor CFAs?
In a single-group cross-sectional model, factor means are fixed at zero. They can be identified only with multiple groups or repeated measures where they must be zero in one group and at one time point.